Job Displacement & Crime Rates

The core of the study's methodology rests on a unique merged longitudinal dataset spanning from 1985 to 2000, encompassing all prime-aged Danish males born between 1945 and 1960. This comprehensive dataset integrates information from five key sources: (i) The Integrated Database for Labor Market Research (IDA), providing annual employer-employee level data including firm identification, workplace details, and employee salaries; (ii) The Central Register of Labour Market Statistics (CRAM), offering weekly unemployment benefit and social assistance records; (iii) The Central Police Register, detailing crime occurrences, charges, convictions, and prison sentences; (iv) Population Registers, supplying demographic and household structure information; and (v) The Danish Student Register, which provides education levels attained by individuals. The linkage across these diverse sources is achieved through anonymized individual Central Person Register (CPR) numbers, a unique identifier present in all datasets. The resulting dataset provides a rich and almost complete picture of employment, unemployment, and criminal activity for the target population. It's crucial to note that weekly unemployment and daily crime data allow precise timing comparisons, enabling a robust analysis of the causal link between job loss and criminal behavior. This detailed data structure forms the foundation for the rigorous investigation of the impact of job displacement on crime rates in Denmark.

The study focuses specifically on job displacement experienced by high-tenure workers during mass layoff events. A mass layoff is defined as a firm experiencing a more than 30% or 40% reduction in its workforce. The threshold is calculated relative to either the firm's peak employment in 1985-1990, its average employment during the same period, or a firm-specific employment trend projected from 1985-1990 data and extrapolated to 1990-2000. This multifaceted definition allows the researchers to account for both sudden and gradual declines in employment, ensuring that the analysis focuses on truly unexpected and impactful job losses. The choice of using a firm-specific employment trend as a benchmark is particularly noteworthy, as it mitigates potential biases arising from gradual firm size changes and allows for isolating genuinely surprising and sudden job displacement events. This precise definition is instrumental in the identification strategy employed, enabling the isolation of the causal impact of mass layoffs on crime rates, independent of other economic fluctuations.

To address potential endogeneity issues, the study employs a rigorous sample selection process. The initial sample, comprising all prime-aged (30-45 years in 1990) native Danish men born between 1945 and 1960, is further refined. The researchers introduce four key criteria: first, individuals must have maintained continuous presence in the dataset throughout 1985-2000; second, the focus is placed on individuals in full-time employment in 1989, minimizing potential biases related to the endogenous nature of job transitions; third, the individuals need to be employed in firms with ten or more workers in 1989, mitigating the influence of disproportionate percentage changes in small firms and excluding self-employed individuals; and fourth, exclusion of individuals enrolled in any educational program between 1987 and 1989. These criteria ensure a homogeneous sample of high-tenure workers with a strong attachment to their respective firms, minimizing the likelihood of confounding factors obscuring the relationship between job displacement and crime. The resulting sample of 102,376 individuals provides a robust basis for examining the causal relationship between job loss and subsequent criminal activity. Focusing on native males of a specific age cohort likely provides a lower bound for the true effect, acknowledging potential limitations in generalizability.

The study's primary finding is a statistically significant increase in the likelihood of criminal activity following job displacement. Specifically, displaced workers show a 0.52 percentage point rise in the probability of committing any crime within the year of their job separation. This overall increase is largely driven by a substantial rise in property crime, with a 0.38 percentage point increase observed – representing approximately 26% of the population-wide average property crime rate. The magnitude of this effect is noteworthy and suggests a considerable impact of job displacement on criminal behavior. Importantly, pre-displacement data reveals no evidence suggesting that individuals more prone to crime were disproportionately selected for layoff, suggesting the observed increase in crime is a direct consequence of job loss, not pre-existing criminal tendencies. This significant increase in crime following job loss aligns with Becker's (1968) economic theory of crime, which posits that criminal behavior is influenced by an individual's assessment of potential benefits and opportunity costs; job loss directly impacts the opportunity costs, increasing the likelihood of criminal activity.

The study emphasizes the significant role of property crime in the observed increase of criminal activity following job displacement. The 0.38 percentage point rise in the probability of committing property crime is a substantial effect, further strengthening the correlation between job loss and criminal behavior. This is in line with Becker's theory, as individuals facing significant financial hardship from job displacement may turn to property crime to address immediate needs. The data strongly supports this notion by showing substantial post-displacement earnings losses experienced by displaced workers. These considerable financial losses create a substantial motivation for engaging in criminal activities. The researchers demonstrate that the post-displacement earnings losses are particularly significant for workers with only a high school education or less, highlighting a disparity in vulnerability to both job loss and subsequent criminal behavior among individuals with different levels of education.

Beyond the immediate impact on the displaced individual, the study explores broader consequences within families and communities. The research shows a moderately increased likelihood of marital dissolution following job displacement, which is a significant life-altering event potentially compounding the negative effects of job loss. Further, while small, the data suggests the existence of intra-family externalities. The displacement of adult family members appears to have a minor impact on the crime rate of younger family members, showing a lagged effect observed a year after the displacement event. These findings point to the far-reaching ramifications of job displacement extending beyond the individual level. Moreover, the impact of displacement on crime is significantly amplified in municipalities characterized by substantial income inequality, indicating that socio-economic context plays a key role in mediating the relationship between job loss and criminal activity.

The study leverages data from 270 Danish municipalities to analyze the relationship between income inequality and crime following job displacement. While Denmark, as a whole, exhibits relatively low income inequality (according to Atkinson & Søgaard 2013, Piketty 2014), the analysis reveals substantial variation across municipalities. The five most unequal municipalities display Gini coefficients approximately double those of the five least unequal municipalities. Notably, in cross-sectional analysis, municipality-level Gini coefficients do not show a significant correlation with overall or property crime rates. However, this analysis is important context for understanding the subsequent findings on the interaction between income inequality and the impact of job displacement on crime. The data used to calculate Gini coefficients come from tax records on wage and capital income, providing a robust measure of income distribution within each municipality. This detailed municipal-level data allows for the investigation of how local income inequality might influence the relationship between job displacement and crime rates.

The research demonstrates that job displacement leads to a notable decline in an individual's income percentile at the municipal level. Specifically, displaced individuals experience a drop of around 2.8 percentile points in the year of displacement, and this effect persists, reaching 3.3 percentile points seven years post-displacement. Critically, the study finds a strong interaction between income inequality and the effect of displacement on crime. Displaced workers residing in municipalities within the upper quartile of the Gini distribution (the most unequal municipalities) are approximately twice as likely to commit a crime post-displacement compared to those in the lower quartile (the least unequal municipalities). This finding highlights a significant interaction effect, suggesting that the impact of job displacement on crime is significantly exacerbated in areas with higher levels of income inequality. The analysis specifically notes that the results are not driven by the Copenhagen city area, suggesting a broader pattern across Danish municipalities.

The study acknowledges potential challenges in interpreting the relationship between income inequality and crime following job displacement. It addresses two key concerns: Firstly, the possibility that income distribution is correlated with municipality-level unobservable factors that also influence crime rates. Secondly, the dynamic nature of income distribution and the possibility that job displacement induces movement of individuals between municipalities with varying income inequality. To mitigate the first concern, the analysis focuses on how changes, rather than levels, in income distribution impact criminal activity. To address the second concern, the researchers investigate whether utilizing current or pre-displacement income distribution measures affects the results. The study also examines the relationship between job displacement and municipal mobility. The findings indicate increased likelihood of moving to another municipality following displacement, but this mobility doesn't significantly confound the estimated relationship between income inequality and the impact of displacement on crime rates. These methodological considerations ensure a robust analysis of how income inequality and geography affect the observed relationship between job loss and crime.

The study reveals that while Denmark has lower incarceration rates and shorter prison sentences compared to the United States, incarceration remains a significant outcome for a substantial portion of convicted individuals. Within the study's sample of Danish males, 26.29% of those convicted received a prison sentence. This highlights the relevance of considering incarceration as a key consequence of criminal activity in this context. Furthermore, the research presents compelling evidence of a strong link between incarceration and substantial earnings losses. Analysis indicates that individuals sentenced to prison experience considerably higher earnings losses than those given alternative sentences such as suspended sentences, fines, or settlements. The difference in earnings losses between those incarcerated and those receiving other sentences is estimated to be as high as 14,000 Danish Kroner (approximately 8% of a standard deviation), highlighting the significant long-term economic consequences of incarceration for displaced workers.

The research carefully considers the multifaceted nature of earnings losses related to incarceration. The study acknowledges that a portion of the observed higher earnings losses among incarcerated individuals is directly attributable to the loss of labor market opportunities during imprisonment. This 'mechanical' impact needs to be differentiated from the 'non-mechanical' impacts of a criminal record, which may persist even after release from prison and affect future employment prospects. Therefore, the analysis focuses on isolating these mechanical and non-mechanical impacts of incarceration to gain a clearer understanding of the full extent of the economic ramifications of criminal activity on earnings. This approach allows for a more accurate assessment of the long-term economic penalties associated with a criminal conviction, particularly in the context of pre-existing job displacement.

The study examines the relationship between job displacement and marital stability, finding a heightened likelihood of marital dissolution following job loss. This aligns with economic models (like Becker, Landes & Michael, 1977) suggesting that reduced earnings following displacement negatively impact both the spouse's outside options and the overall value of the marital relationship. The research uses the multiple years following displacement to observe how the probability of divorce evolves in the short, medium, and long term. While the increased probability of divorce is noted, the main focus of the study is on the interaction between marital status and the increased propensity toward crime after job displacement. The analysis considers whether being married or not in 1989 and the current marital status impact the effect of displacement on the likelihood of committing crime. Results show a larger effect on crime for individuals not married in 1989, but the overall effect of displacement on crime isn't significantly altered by including current marital status as a control, suggesting that marital changes might not be the primary mechanism explaining the increased crime rates. The data, therefore, offers additional insight into the social and economic consequences of job displacement beyond the direct impact on the individual.

The research explores the potential intergenerational consequences of parental job displacement, investigating whether fathers' job loss affects the subsequent criminal activity of their children. The study leverages the longitudinal dataset to assess the impact of displacement events experienced by parents (from earlier birth cohorts) on the criminal activity of their children (from later birth cohorts). This detailed data allows for the examination of intergenerational spillovers. The analysis considers various potential mechanisms linking parental job loss to children's crime, including reduced family income and lower educational attainment among children of displaced fathers (as found in prior research by Oreopoulos, Page & Stevens, 2008, and Stevens & Schaller, 2011, respectively). While a correlation between parental job loss and youth crime is acknowledged from existing literature, this study specifically utilizes a micro-panel design to isolate these effects, carefully matching displacement events to subsequent criminal activity of younger family members, providing unique insights into the intergenerational transmission of the negative consequences of job displacement. The findings suggest some impact, but it is relatively small compared to the direct impact on the displaced parent.

The analysis investigates how family structure might influence the relationship between job displacement and crime. The research specifically examines whether the impact of job loss on crime operates through changes in spousal labor supply. By comparing earnings losses at both the individual and family levels, the study finds that the loss of earnings following displacement is similar regardless of whether the analysis is conducted at the individual or family level. This suggests spousal income remains relatively stable after the male adult experiences displacement. This indicates that spouses might engage in income pooling or other strategies to smooth consumption following a job loss. This consistent finding across individual and family level analyses suggests that changes in spousal labor supply are unlikely to be the primary factor driving the increased likelihood of crime following job displacement. The relatively stable spousal income suggests the presence of mechanisms within the family that mitigate some of the immediate financial shocks associated with job loss for the displaced individual.